William Herschel, Flamsteed Numbers and Harris's Star
Total Page：16
File Type：pdf, Size：1020Kb
JHA, Vol 45(3), 2014 WILLIAM HERSCHEL, FLAMSTEED NUMBERS AND HARRIS’S STAR MAPS WOLFGANG STEINICKE, Umkirch, Germany Abstract The paper discusses the origin of the commonly used Flamsteed star numbers, which are missing in John Flamsteed’s British Catalogue (1725) and accompanying Atlas coelestis (1729). Johann Elert Bode introduced these numbers in the Sammlung Astronomischer Tafeln (1776). Herschel independently created them in 1781 for his third star review. At that time he purchased Flamsteed’s Atlas coelestis and the question arises, which star chart did he use before, especially when Uranus was discovered. This leads to two star maps, linked to the Welsh astronomer Joseph Harris and based on stars whose positions were measured by Flamsteed and Halley. Key words John Flamsteed, Edmond Halley, Johann Elert Bode, Joseph Harris, British Catalogue, Atlas coelestis, star reviews, double stars, Uranus Flamsteed and Halley: Perfectionism and Piracy Astronomers are familiar with star designations such as 19 Orionis, going back to John Flamsteed (1646–1719), the first Astronomer Royal at Greenwich. This star is better known as Rigel or β Orionis. Though star names are old, Greek letters were not introduced until 1603, by Johann Bayer in his famous star atlas Uranometria, based on a catalogue of nearly 1200 stars.1 A century later, Flamsteed measured the positions of some 3500 stars with a mural arc and sextant. Unfortunately, during his lifetime he published neither his catalogue nor the related star charts, which did not appear until 1725 and 1729 respectively. Surprisingly, neither publication contains Flamsteed numbers! So, who numbered the stars and when? To solve the mystery one must go back to Edmond Halley (1656–1742). Pressured by Isaac Newton, President of the Royal Society, Halley edited an unauthor- ized version of Flamsteed’s catalogue in 1712, when 100 copies were distributed. The catalogue lists 2866 stars, ordered by constellation. In each constellation they are arranged by increasing right ascension — and grouped in units of 5. The numbers are given in the first column, headed “Britain”.2 Flamsteed was angry about the pirated edition, not only because Halley had stolen his data, but even more because his work was still unfinished. Flamsteed the perfection- ist feared to be linked with incorrect data. A few years later he gained control of 300 as yet undistributed copies and burned the parts he did not like, in front of the observatory. Pages consigned to the flames included the catalogue and the spurious part of the work, which professed to be his observations made with the mural arc.3 However, for future use he saved at least one copy of Halley’s pirated edition, consisting of those parts otherwise burned (mainly the catalogue). A copy — containing unique annotations — is in the Copyright © The Author(s) 2014 288 Wolfgang Steinicke possession of Owen Gingerich;4 these annotations were made by Flamsteed’s assistant Joseph Crosthwait,5 of course following instructions of the Astronomer Royal. They show that some stars should be moved from one constellation to another and, interestingly, the column “Britain” is crossed out (Figure 1). For his edition, Halley used a manuscript of Flamsteed, written in 1707 (and now in Cambridge University Library). It shows the stars grouped in units of 3 and numbered in units of 5. This finding led to the interpretation that it was Flamsteed who invented the (preliminary) numbers — and not Halley.6 But this must be doubted for two reasons:7 (1) the different grouping, and (2) the fact that the first column, presenting the numbers in units of 5, seems to be added later (in a slightly different handwriting). The annotations could have been made by Halley when preparing the 1712 catalogue. In June 1716 Flamsteed sent the rescued part of Halley’s catalogue (with Crosthwait’s annotations) to Abraham Sharp (1653–1742); Sharp had worked for Flamsteed until 1700 and later was involved in the final publication of the catalogue. The handwritten corrections clearly do not represent Flamsteed’s final disposition of the stars (Figure 2). He still had a few years to rethink which constellation the observed stars belonged to, and in particular he decided to include the new constellations introduced by Hevelius. The perfectionistic Flamsteed took his time to fine-tune his catalogue, with the result that the astronomical community had to wait until 1725 — six years after his death! — for the official edition: the Stellarum inerrantium Catalogus Britannicus ad annum Christi completum 1689, known as the British Catalogue. The star catalogue is con- tained in the third of three volumes, published as Historia coelestis Britannica (vols i and ii give the observational data, dating from 1675–89 and 1689–1720 respectively). The British Catalogue is the basis of Flamsteed’s Atlas coelestis; this influential star atlas appeared in 1729.8 If one compares the published British Catalogue with Halley’s 1712 version, there are some remarkable differences. (1) There are no star numbers. (2) The constellations are ordered in a different way. (3) Some stars have been shifted from one constella- tion to another. (4) 53 stars have been added to those in the first two volumes of the Historia coelestis, so we now have 2919 stars; this changes the star order in the affected constellations. We can only speculate why Flamsteed did not use star numbers. Probably he was not satisfied with his catalogue right up to his death in 1719 and so he refrained from a final numbering — or perhaps he never wanted a numbering, and his wishes were respected by the editors of the British Catalogue. But the commonly used Flamsteed numbers are based on the British Catalogue, and Halley’s pirate edition is not the source. Though Halley introduced a “natural” numbering system, it was only preliminary. So the question remains, who introduced them and when they were published? A possible choice is John Bevis (1695–1771), the creator of the “forgotten” star atlas Uranographia Britannica.9 It is known that the star catalogue on which it is based, made about 1750, contains Flamsteed numbers. Bevis obviously was inspired by Halley. Unfortunately, this work was never published (only the plates were printed, but they do not show numbers). William Herschel, Flamsteed Numbers and Harris’s Star Maps 289 FIG. 1. Page of Halley’s 1712 unauthorized edition of Flamsteed’s star catalogue, showing Lyra and the first stars of Cygnus. The stars are numbered in the column “Britain”. The marks in the right margin, made by Flamsteed’s assistant, Joseph Crosthwait, show where stars were moved. Stars a and b went from Cygnus (below) to Lyra (above) and two to Vulpecula (compare Fig. 2). The fainter crossed lines cancel Halley’s descriptions and the numbers that Flamsteed did not like in the left-most column. Collection of Owen Gingerich. 290 Wolfgang Steinicke FIG. 2. Tables of Flamsteed’s British catalogue, published 1725, showing Lyra and the first stars of Cygnus (now on different pages). The Cygnus stars a and b (see Fig. 1) are now in Lyra: a = 13 (π) Lyr, b = 16 (ρ) Lyr. Star c is now 1 (κ) Cyg (note that the right ascension arcminute was corrected from 27 to 28). Star d should move to Lyra, but is still in Cygnus (4 Cyg). The two stars in Fig. 1 moved to Vulpecula became 2 Vul and 10 Vul (page not shown). The Role of Johann Elert Bode On the basis of research by Deborah Warner, it has been widely accepted that Flamsteed numbers were introduced by Joseph Jérôme Lalande in his revision of the British Catalogue, published 1783 in the French almanac Éphémérides des mouvemens célestes.10 But this is not the earliest source, for the eminent Berlin populariser of William Herschel, Flamsteed Numbers and Harris’s Star Maps 291 astronomy Johann Elert Bode (1747–1826) had presented them as early as 1776.11 The numbers are contained in the Sammlung Astronomischer Tafeln, a three-volume collection of astronomical tables, made in partnership with Johann Heinrich Lambert and Johann Karl Gottlieb Schulze. Vol. i contains an 89-page star catalogue, compiled by Bode.12 This “Stern-Verzeichnis nach vier Astronomen” assembles the star catalogues of Hevelius, Flamsteed, Lacaille and Bradley.13 Concerning Flamsteed, Bode wrote: “Flamsteads [sic] vollständiges Verzeichnis, welches sich in seiner Historia Coelestis 3ten Bandes der 2ten Auflage von 1725 befindet [Flamsteed’s complete catalogue which is in vol. 3 of the second edition of the Historia Coelestis of 1725].” When Bode speaks of the second edition, he must regard Halley’s publication of 1712 as the first. The catalogue lists 3175 stars. They are arranged by constellation, starting with Cassiopeia (see Figure 3). The first column, headed “Fl. No”, gives the correct Flamsteed number. The positions (right ascension, declination) are calculated for 1800. Thus Bode’s star catalogue of 1776 is an extended and updated version of the British Catalogue. Bode was to use the Flamsteed numbers in his new almanac, Astronomisches Jahrbuch. It regularly contained a table, titled “Verzeichnis der 280 der vornehmsten Fixsterne [List of the 280 most prominent fixed stars]”. In the first two issues, for the years 1778 and 1779 (published in 1776 and 1777), the table does not show these numbers, but the third almanac, Astronomisches Jahrbuch für 1781 (published in 1778), does give them.14 The first column is headed “No. nach Flamsteed” (the table starts with “88 | Algenib im Pegasus”). Strangely, the Flamsteed column is omitted from 1781 (Jahrbuch for 1784) onwards. In 1782 Bode published his popular star atlas Vorstellung der Gestirne.15 The second part of the introduction presents the stellar basis: a catalogue of 5058 stars, covering 37 pages.